TiHabanero

I had the rear end set up perfectly for 135mm axle after brazing to the main frame. Next brazed in the brake bridge and fender mount (chain stay bridge) and the spacing moved from 135 to 130. Not sure how that happened as the pieces were formed to fit perfectly at the 135 spacing. How did the rear end pull together?

unterhausen

it's a pain, isn't it? You can try witch wanding it back, heat the outside of the stays, not too much heat. When I put bridges on, I finish up by heating the outside of the stays, but you can do it afterwards.

I found a video once where a builder was aligning a frame using witch wanding. It was pretty amazing. Tried to find it recently and failed.

Andrew R Stewart 

This is why many will jig the rear end at a few MMs wider then the final spec. Start looking down the lengths of built bike's seat stays and you'll see many with a slight dog leg centered at the brake bridge.

Basically the concentration of heat over time on the stay's inside will initially cause that side to expand (and the bridge can slip out of place if not secured well) then on cooling contract. It will contract more then it expanded thus pulling the stay in and resulting in a narrower drop out width. Andy

dsaul

It happens every time. Less with brazing than with TIG, but I always wedge the bridge in tight and try to do the acute angle of the joint last, so the shrinking of the filler material pulls the stays apart again. I usually have to do some cold setting of the stays after the bridge goes in, even with using the dummy axles with 2mm extra width.

unterhausen

Never thought about doing the different sides in order. I'm sure I always do the acute side last just because. Now I wonder if doing the acute side first would be better. Trying to think about how to do an experiement without wasting a lot of tubing

Doug Fattic 

One of the biggest learning curves is knowing how to compensate for heat distortion caused when brazing a joint. Many classic European builders just bent things back. Reynolds 531 tubing was popular because it could be altered after brazing. When heat-treated 753 came along that was no longer a suitable solution and more thought and care had to go into procedures and sequences.

One of the most common ways to end up with 130 rear spacing is to start brazing your bridges in when they are wider. For example if you discovered your dropouts ended up at 126 you know you need to start at 134. Actually you want a little bit of clearance when you are finished so begin with 134.5. Another possible solution is not to focus your flame entirely on the inside but also heat the outside before and after brazing in the bridges. This is obvious but lower temperature silver does not distort as much as brass. Of course the miters have to be more perfect.

unterhausen

I am curious what the people that build with TIG do. Certainly there are a number of builders that have some method that doesn't involve bending the stays.

Heating the outside of the stays when I'm done brazing the bridge has been working for me. I would really like to understand the mechanism that causes this problem. I fillet brazed a dishing gauge in the EVT style and the center tube had horrible distortion. Never seen anything like it on a fillet brazed frame. I did get it to be a lot rounder by selective heating of the tube.

TiHabanero

This is the first time the drop outs have closed up when attaching the chain stay brace and brake bridge. Perhaps it is due to using bronze filler instead of silver as I did in the past?

Andrew R Stewart 

Good chance yes. The higher heat of brass will potentially distort things more. Andy

unterhausen

Did you try heating the outside of the stays at the bridges? I have some s-bend stays that don't quite curve the way I want and I was thinking about trying to change them using that method. If I do that, I'll take pics.

TiHabanero

The stays are 47cm so I simply cold set them to spec. Next time I will hit with heat and see how it goes.